MACD Histogram
Last Updated on November 24, 2020 by Oddmund Groette
I’m trading some strategies based on mean reversion on a sample of 77 ETFs. Those ETFs are the most liquid ones. One of the strategies I trade is based on MACD-Histogram. It’s a very much used indicator I have tweaked a little bit. On a per trade basis, this is one of the best strategies I have. Here you can read about the indicator. I have tweaked this one a little bit to the following:
- The MACD Histogram bar must have fallen 4 days in a row.
- The fourth latest bar must have been below zero.
- The current close of the ETF must be lower than the day before.
Entry is on the close. The exit is on the first day when the close is higher than the day before. Here is an example (exit was on the close the day after because the ETF rose in price):
For short it is vice versa. However, long is a lot better than short. In general short is a lot more difficult to trade.
In total, this strategy has generated 6669 trades on my portfolio of 77 ETFs.
This diagram shows the average gain per trade since 2000. Among those 77 ETFs hardly any is negative:
From month to month we get the following bar chart:
I think this is a pretty good result. Assuming one can only trade a maximum of ten positions at a time as a portfolio, we get this equity curve:
This strategy is in cash most of the time! This strategy can be one of several if swing trading. This proves that a very simple strategy can be very efficient.
Very interesting – I have programmed these into ninjatrader for automatic execution – The backtesting looks good.
I am looking for a similar intraday strategy – example – if an ETF reverts from days low/high – weeks low/high, months high/low, years low/high – 200ma – price where a lot of trading has been done – or similar. If condition met, then hold the trade until close. I will look into this when I have time. It could be an interesting automated intraday strategy.
Truls.
Hi, you programmed the same strategy as I made? Actually, throwing out the bond/high yield/interest rates ETFs the results are better. Works very good on stock ETFs. The only ETFs which are bad are typical commodity,oil etc. I have tested this on OBX in Norways as well – not good (and to be expected).
As regards your thoughts about intraday strategy: nice if you could give some feedback when you’re done.
What kind of MA is used? I don’t like indicators where I have to adjust parameters…. but the test looks nice 🙂
The “standard” MA is used. I can’t remember, but think it is 9 and 26.
It is quite easy to make automated trading systems in ninjatrader, so I just put in your criterias and I am currently testing it in my sim account before putting it on my live account,
Truls
Thanks for sharing this strategy. It fits my trading style really well. I have done lots of my own back testing using your entry rules but have tried various different exits. All produce nice results.
I am a fan of using an extreme over sold level of RSI 2 to enter long but from my back testing, this method using MACD H seems to produce better results.
It also works well with weekly charts too.